4716
Ind. Eng. Chem. Res. 2001, 40, 4716-4723
APPLIED CHEMISTRY Tertiary Resource Recovery from Waste Polymers via Pyrolysis: Neat and Binary Mixture Reactions of Polypropylene and Polystyrene Hsi-Wu Wong and Linda J. Broadbelt* Department of Chemical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208
Batch pyrolysis of polypropylene, polystyrene, and their binary mixture was carried out at temperatures of 350 and 420 °C. Two different loadings were also studied for neat polypropylene and polystyrene to assess the impact of total pressure, which was also a function of reaction time and temperature. For neat polypropylene pyrolysis, total conversion was an order of magnitude higher at 420 °C than at 350 °C. For neat polystyrene pyrolysis, conversion reached approximately 75% at 350 °C, whereas at 420 °C, the conversion reached a maximum of approximately 90% after 10 min of reaction and decreased to approximately 70% after 180 min. For binary mixture pyrolysis, the overall conversion was higher than the average of the conversions for the two neat cases. The conversion of polystyrene remained the same, but significant enhancement of the polypropylene conversion was observed. These results suggest that the more facile degradation of polystyrene helped to initiate the less reactive polypropylene. Introduction In recent years, post-consumer wastes have caused increased concern because of the escalation of municipal solid wastes (MSWs) generated. Plastics make up a significant portion of MSWs. In the United States alone, over 21.5 million tons of plastics are manufactured annually, and production is projected to increase to 26.7 million tons in 2005.1 In 1997, 9.9 wt % of MSWs comprised plastics, which was more than 20 vol %.2 Within the plastic waste stream, 13.0 wt % was composed of polypropylene (PP), and 9.8 wt % contained polystyrene (PS).1 Tertiary recycling, in which discarded plastic products are converted into high-value petrochemical or fuel feedstocks, has received increased attention in recent years. Two methods of tertiary recycling, chemical recycling and thermal recycling, can be applied, depending on the polymer involved. Condensation polymers, such as polyesters, nylons, and polyurethanes, can undergo decomposition via chemical methods such as glycolysis, methanolysis, and hydrolysis. Addition polymers, such as polystyrene, polyethylene, and polypropylene, are typically broken via thermal or catalytic cracking.3 Thermal degradation is the simplest form of tertiary recycling of addition polymers, and therefore, it has been studied most extensively. One of the biggest costs associated with polymer recycling is the sorting of the original polymers. However, the conditions required to break down addition polymers lead to a broad product distribution even for a single component. Therefore, tertiary recycling of * Author to whom correspondence should be addressed. Phone 847-491-5351; Fax 847-491-3728; E-mail broadbelt@ northwestern.edu.
addition polymers via processing of a multicomponent polymeric mixture is one potential solution to eliminate the cost of sorting. The economics of fuel or chemical production from a mixed stream of addition polymers can be further improved by optimizing the coprocessing conditions. Our objective was to investigate the interactions of different polymers during pyrolysis to determine whether any synergistic effects are present during coprocessing. Specifically, we conducted binary mixture pyrolysis of polypropylene and polystyrene as a model of mixed plastic waste and compared the results to those obtained for the neat pyrolysis of the individual components. Although the neat pyrolyses of both polypropylene4-8 and polystyrene9-13 have been extensively studied and a small number of investigations have been carried out on the pyrolysis of mixed polymers,14-18 very few researchers have focused on the binary system of polypropylene and polystyrene.18 Furthermore, our investigation placed emphasis on a detailed quantification of the low-molecular-weight products, which would be useful as fuels and chemicals. Experimental Section Batch pyrolysis experiments were carried out by loading either polystyrene or polypropylene into a 3.1mL glass ampule (Wheaton). This simple reactor configuration facilitated product analysis and allowed interactions between the volatile products and the degrading polymer, as would occur in a continuous-flow reactor, to be gauged. Two different loadings, 20 and 10 mg, were studied. For binary reactions, 10 mg of each polymer was used. The polystyrene (Mw ) 111 800, Mn ) 98 100) was prepared in our laboratory by anionic polymerization.16 Polystyrene synthesized by anionic
10.1021/ie010171s CCC: $20.00 © 2001 American Chemical Society Published on Web 09/29/2001
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001 4717
polymerization does not contain weak links, and therefore, its degradation behavior can be attributed to the regular polystyrene structure. Furthermore, this enabled comparison with our previous work.16,19,20 The polypropylene was obtained from Aldrich (syndiotactic; Mw ) 127 000, Mn ) 54 000) in powder form. After being purged with argon for 2 min, each ampule was sealed using an oxygen/propane flame; then, the sample was reacted in an isothermal fluidized sand bath. Two reaction temperatures, 350 and 420 °C, were studied. A temperature of 350 °C was chosen because it is above the ceiling temperature of polystyrene but below that of polypropylene, and 420 °C was chosen to obtain appreciable conversion of polypropylene and because it was the same temperature used to study the degradation of polyethylene.21 The reaction times ranged from 1 to 180 min. These times allowed a large range of conversions to be studied for both polymers. In addition, runs with a reaction time of 18 h for polypropylene pyrolysis at 420 °C were carried out. Upon completion of the reaction, the ampule was removed from the sand bath and quenched in another sand bath set at ambient temperature. At least two replicates, and in most cases three replicates, were performed for each reaction time. Products that were gaseous at room temperature from neat polypropylene and binary pyrolysis were analyzed by placing the ampule inside a 53-mL flask with a Tygon tube on one end and an injection port on the other. Both ends were then sealed with septa. The flask was purged with helium for 10 min, and after the ampule was broken, the gases were allowed to equilibrate for 30 min. Gas samples were taken using a gastight syringe and were then identified and quantified using a HewlettPackard 5890 Series II Plus gas chromatograph equipped with a thermal conductivity detector (TCD) and a 6-ft stainless steel Porapak Q column (Supelco). Liquid and solid products were extracted with 1.5 mL of HPLC-grade methylene chloride. The product solution was first passed through a 0.45-µm polypropylene filter (Alltech) attached to a syringe and then through a Waters gel permeation chromatograph (GPC). The molecular weight distribution of the soluble polymer fraction was measured. Polystyrene and its degradation products were completely soluble in methylene chloride, whereas only those polypropylene-derived products up to C25 were completely soluble. Products with molecular weights less than ∼400 g/mol were collected using a fraction collector attached to the GPC outlet. An external standard (biphenyl) was added after fraction collection. Product identification and quantification were achieved using a Hewlett-Packard 6890 Series gas chromatograph-mass spectrometer and a Hewlett-Packard 6890 Series gas chromatograph with a flame ionization detector (FID), each equipped with a HewlettPackard 30-m cross-linked 5% phenyl methyl silicone capillary column. The percent conversion of the polymer, X, was defined according to eq 1
X)
W × 100% Wo
(1)
where W is the weight of the products analyzed by gas chromatography, which is equal to the sum of the weights of the gaseous products (Wg) and the liquid products (Wl), and Wo is the initial weight loading of polymer. More specifically, Wl is the sum of the weights
Figure 1. Total conversion of neat polypropylene pyrolyses at 350 and 420 °C.
of nonaromatic liquid products with carbon numbers less than or equal to 25 and aromatic products with molecular weight less than or equal to 404. In the case of neat polystyrene pyrolysis, less than 1% gaseous products were found even at 420 °C for 180 min of reaction, and thus, Wg was neglected. For binary mixture pyrolysis, the conversion of polystyrene was calculated from the weight of all aromatic products, and the weight of all other products was used to calculate the conversion of polypropylene. The selectivity, S, of a certain species A was determined on the basis of eq 2
S)
WA × 100% W
(2)
where WA is the weight of species A. Product yields used for neat polypropylene and binary pyrolysis were normalized by dividing the number of millimoles (mmol) of each product by the initial molar loading of polymer repeat units. Finally, the error bars shown in the figures represent the standard deviations of experiments that have been at least duplicated. Results and Discussion 3.1. Neat Polypropylene Pyrolysis. The overall conversion of polypropylene during neat pyrolysis increased with respect to reaction time for both temperatures and loadings, as illustrated in Figure 1. After 90 min of reaction, the conversion at 420 °C reached approximately 60%, and no significant change was observed at longer reaction times. This suggests that little additional conversion can be achieved even for very long reaction times at this temperature in a closed batch reactor of 3.1-mL volume. This was verified by running the reaction for 18 h, as the conversions were 65.2 and 59.0% for reactant loadings of 10 and 20 mg, respectively. The conversion was significantly lower at 350 °C, achieving a maximum value of only around 3% after 180 min of reaction. The pressure increased by less than 0.1 atm for both loadings. Clearly, as shown in Figure 1, the effect of temperature on polypropylene pyrolysis is marked, whereas the influence of different loadings is not significant. For 420 °C runs, the conversion of the sample with a loading of 10 mg reached 57.5% after 180 min of reaction, while the conversion for the 20 mg loading was slightly higher at 62.7%. However, after 18 h of reaction, the conversion for the 10-mg loading was
4718
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001
Figure 2. Weight yields of different product fractions as functions of reaction time: (a) 420 and (b) 350 °C.
higher than the conversion for the 20-mg loading, as described above. The slight differences in conversion between the two loadings can be explained by the differences in total pressure between the two systems. Whereas the initial pressure was 2.3 atm for both loadings, the pressure was approximately 5.8 atm for the higher loading and 3.9 atm for the lower loading after 180 min of reaction. The higher pressure for the 20-mg sample results in an enhancement in bimolecular reactions and a faster rate of conversion. However, the final conversion is dictated by thermodynamics and is higher for the system with the lower total pressure. If the products were divided into three fractions, C1C4, C5-C10, and C11-C25, the yields of these fractions would show different temporal behaviors, as shown in Figure 2. At 420 °C, the yield of the C1-C4 fraction increased monotonically with reaction time, whereas the C5-C10 fraction reached a subtle maximum around 120 min with no significant change observed at 180 min. Finally, the C11-C25 fraction reached a distinct maximum around 60 min. No significant difference between the two loadings was observed at this temperature. At 350 °C, all three fractions increased with reaction time, with the yield of the C1-C4 fraction significantly lower than the yields of the other two fractions. Slightly higher yields were found for a loading of 10 mg for the C5-C10 and C11-C25 fractions at reaction times of 120 and 180 min. The results suggest that, as the reaction time and temperature increase, the heavier products decompose into lighter ones. The reaction products of polypropylene pyrolysis consist of four major categoriessalkanes, alkenes, dienes,
Figure 3. Representative normalized yields of (a) alkanes and (b) alkenes from polypropylene pyrolysis at 420 °C for a 20-mg loading.
and aromatic compounds. The yields of alkanes and alkenes are plotted in Figure 3. Lower-molecular-weight species were found in higher yields, whereas the yields of compounds with carbon numbers greater than 10 were notably lower. For alkanes, ethane (C2), n-pentane (C5), 4-methylheptane (C8), C11, C14, and C17 were found in high yields. The alkanes were thus dominated by products with carbon numbers of C3n-1, with n ) 1, 2, 3, 4, .... The yields of the majority of the alkanes increased with respect to reaction time. For alkenes, propylene oligomers (C3, C6, C9, C12, C15, C18, C21, and C24), i.e., C3n, n ) 1, 2, 3, 4, ..., were the dominant products. It is believed that commercial syndiotactic polypropylene made using coordination catalysts would have chains with double bonds on the ends,22 such that the penpenultimate bonds from the chain ends would be weaker than normal carbon-carbon bonds. In addition, olefinic end groups are produced during degradation. Therefore, isobutylene (C4 alkenes), which can be formed from polypropylene chains with double bonds on the ends, was found in high yield. Dienes, which were found beginning with C7 and were present in relatively low yields, appeared as C7, C10, C13, C16, C19, C22, and C25, i.e., C3n-2, n ) 3, 4, 5, 6, .... Finally, aromatic compounds were also found as minor products. Their yields were comparable in magnitude to the diene yields. As noted above, the product distribution revealed that the dominant alkenes were for compounds of the form
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001 4719
Figure 4. (a) Formation of two kinds of polymer chains (I and II) through bond fission, hydrogen abstraction, and β-scission and proposed reaction mechanisms of polymer chains (b) I and (c) II.
C3n, whereas alkanes and dienes appeared in the forms C3n-1 and C3n-2, respectively. This product distribution is in agreement with observations reported in the literature5,6,8,23 and can be explained by the mechanism illustrated in Figure 4, which is based on the reaction types proposed by Tsuchiya et al.5 We have clarified the mechanistic picture by delineating the pathways that lead to the different observed product classes (alkenes, alkanes, and dienes) with the dominant carbon numbers (C3n, C3n-1, and C3n-2, respectively). The initiation step of the free-radical mechanism is simply to break any of the polypropylene chains into two shorter end-chain radicals. The end-chain radicals (or midchain radicals formed subsequently) can abstract hydrogen from a long polypropylene chain to form a tertiary radical, as shown in Figure 4a. Upon undergoing β-scission, the tertiary polymer radical is broken into two fragments, one with a double bond on the end (denoted by I) and the other with a secondary free radical (denoted by II). When polymer chain I is attacked by another free radical and β-scission occurs, dienes (in the form of C3n-2) and alkenes (in the form of C3n) can be formed, as shown in Figure 4b. On the other hand, polymer chain II can undergo three kinds of reactions such that alkanes (in the form of C3n-1) and alkenes (in the form of C3n) can be formed, as shown in Figure 4c. Although other steps not explicitly drawn are possible, the formation of C3n alkenes, C3n-1 alkanes, and C3n-2 dienes as the dominant products suggests that the mechanism in Figure 4 captures the major reaction pathways. 3.2. Neat Polystyrene Pyrolysis. The overall conversions for the neat pyrolysis of polystyrene at both temperatures and loadings are shown in Figure 5. For high-temperature (420 °C) reactions, polystyrene degraded completely in 10 min, and reactions occurred to form condensed aromatic products that were greater than 404 g/mol and thus were not included in the calculation of conversion. These products were sug-
Figure 5. Total conversions of neat polystyrene pyrolyses at 350 and 420 °C.
gested by the appearance of a new peak on the gel permeation chromatograms and the increasing yields of naphthalene and 2-phenylmethylnaphthalene after 10 min of reaction (not shown). For low-temperature (350 °C) reactions, the conversion reached approximately 75% after 180 min of reaction. No significant influence of different loadings on the total conversion at either temperature was observed. The major products of neat polystyrene pyrolysis were a function of temperature and were similar to those reported in the literature.10,11,13 The mechanism of polystyrene decomposition leading to the major products is depicted in Figure 1 of ref 20. At 420 °C, the major products for polystyrene pyrolysis were toluene, ethylbenzene, and styrene. As shown in Figure 6, the selectivity of styrene was around 70% initially and went down as the reaction time increased. However, the selectivities of toluene and ethylbenzene increased monotonically. One possibility is that styrene began to decompose or react with other heavier compounds to
4720
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001
Figure 6. Selectivities of toluene, ethylbenzene, and styrene during polystyrene pyrolysis at 420 °C.
Figure 8. Comparison of conversion for pyrolysis of a binary mixture of PS and PP to that for neat pyrolysis at 420 °C.
Figure 7. Selectivities of styrene and 2,4-diphenyl-1-butene during polystyrene pyrolysis at 350 °C.
Figure 9. Comparison of conversion for pyrolysis of a binary mixture of PS and PP to that for neat pyrolysis at 350 °C.
form toluene and ethylbenzene. The selectivity of styrene for the 10-mg loading was slightly higher than that for the 20-mg loading, whereas the selectivities of toluene and ethylbenzene were slightly lower with decreasing loading. This behavior is consistent with enhanced decomposition of styrene to form toluene and ethylbenzene at higher total pressures. The pressure after 10 min of reaction was around 5.1 atm for the higher loading and 3.7 atm for the lower loading. At 350 °C, the major products of polystyrene pyrolysis were styrene and 2,4-diphenyl-1-butene (styrene dimer). As shown in Figure 7, the selectivities of the two compounds were relatively insensitive to reaction time for the 10-mg loading, whereas they decreased slightly with respect to reaction time for the 20-mg loading. As was observed at 420 °C, the secondary reactions of the olefinic compounds were enhanced at higher total reaction pressures. This behavior is more marked at longer reaction times, when the pressure difference between the two loadings is more severe. After 180 min of reaction, the pressure was near 4 atm for the higher loading and 3.1 atm for the lower loading, whereas the initial pressure for both loadings was 2.1 atm. 3.3. Pyrolysis of a Binary Mixture of Polypropylene and Polystyrene. The total conversion during binary mixture pyrolyses at 420 and 350 °C are compared to the conversions of neat polystyrene and polypropylene in Figures 8 and 9 for 420 and 350 °C, respectively. For both temperatures studied, the overall conversion for binary mixture pyrolysis was higher than the average of the neat cases represented by the dashed lines in the figures, which indicate the conversion if no
Figure 10. Comparison of PS conversion during binary mixture pyrolysis to that obtained for neat pyrolysis.
interactions occurred between the two polymers. By attributing all aromatic products formed during binary mixture pyrolysis to polystyrene, it was observed that the conversion of polystyrene in the binary reactions was similar to that in the neat cases at both temperatures, as shown in Figure 10. In addition, the molecular weight distributions of polystyrene during neat pyrolysis at both loadings and binary mixture pyrolysis were also similar at 350 °C (Figure 11). However, the conversion of polypropylene was significantly enhanced during the binary reactions, as shown in Figure 12. Although performing high-temperature GPC on all of the polypropylene samples was prohibitive, analysis of two different reaction times revealed that the molecular weight of PP
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001 4721 Table 1. Representative Experimental Data for PP/PS Binary Mixture Pyrolysis: (a) Alkanes, (b) Alkenes, and (c) Dienes reaction temp (°C) reaction time (min) total conversion (%)
420 10 68.60
carbon number
Figure 11. Changes in the number-average (Mn) and weightaverage (Mw) molecular weights of polystyrene during pyrolysis at 350 °C as functions of time.
Figure 12. Comparison of polypropylene conversion during binary mixture pyrolysis to that obtained for neat cases. Only the 20-mg loading is shown because the neat conversion was insensitive to loading.
was lowest for the binary mixture sample. Furthermore, rheological measurements were also qualitatively consistent with the GPC data. The normalized yields of polypropylene-derived compounds for representative binary mixture reaction conditions are listed in Table 1. For reactions at 350 °C, the yields of both alkanes and alkenes were higher for binary reactions than for neat pyrolysis. At 420 °C, the alkane yields were slightly lower in the binary reaction, whereas the alkene yields were higher, contributing to the higher total conversion for the binary reactions. To understand the origin of the enhancement of polypropylene’s reactivity in the presence of polystyrene, the kinetic coupling formulation developed by LaMarca et al. for binary mixture pyrolysis was used.24 Although their analysis was restricted to small molecules, their approach provided initial quantitative insights into the interactions between the two polymers. The enhancement or retardation of the rate of disappearance of a component in the coupled system compared to that observed in the neat degradation was predicted by calculating the rate enhancement, which is defined as the rate of disappearance of a given component in the mixture relative to its neat degradation rate. If the rate enhancement for component i is equal to 1, then the addition of the second component has no impact on the rate of decomposition of i. If rate enhancements for both components are greater than 1, this is a clear indication of mutual synergistic effects. To obtain the rates of
420 90 70.76
420 180 72.73
350 10 1.80
350 90 31.06
350 180 46.08
normalized yield (mmol/mol)
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21 C22 C23 C24 C25
1.26 2.80 0.73 0.34 12.66 0.21 0.00 1.99 0.09 0.00 1.33 0.00 0.00 0.15 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00
(a) Alkanes 15.55 31.86 22.89 36.78 11.30 20.45 4.01 7.01 28.88 34.83 1.73 2.48 0.00 0.00 2.53 3.41 0.21 0.38 0.00 0.00 2.80 2.61 0.00 0.00 0.00 0.00 0.28 0.16 0.00 0.00 0.00 0.00 0.10 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21 C22 C23 C24 C25
1.36 22.93 5.25 0.88 13.62 1.65 1.93 31.74 1.58 0.73 6.05 3.96 1.91 10.17 0.00 0.37 3.77 0.00 0.00 2.21 0.00 0.00 0.50 0.00
(b) Alkenes 7.75 8.91 68.23 70.72 39.16 52.34 6.13 8.24 27.03 29.42 3.81 3.54 3.52 3.88 29.25 24.49 12.03 12.25 1.87 1.09 5.77 3.84 2.53 1.24 5.51 4.42 9.09 6.56 0.00 0.00 1.38 1.66 3.28 2.03 0.00 0.00 0.00 0.00 0.50 0.15 0.00 0.00 0.00 0.00 0.16 0.05 0.00 0.00
0.00 0.18 0.00 0.00 0.39 0.00 0.00 0.59 0.84 0.00 0.15 0.00 0.00 0.12 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.09 1.89 0.25 0.00 2.46 0.00 0.00 5.22 0.54 0.00 1.49 0.48 0.00 2.19 0.00 0.16 0.39 0.00 0.00 0.49 0.00 0.00 0.14 0.00
0.19 4.03 0.77 0.23 6.36 0.25 0.00 10.19 0.21 0.00 3.18 1.74 0.21 4.55 0.00 0.29 1.28 0.00 0.00 1.43 0.00 0.00 0.25 0.00
C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21 C22 C23 C24 C25
0.79 0.00 0.19 0.10 0.00 0.00 2.65 0.00 0.00 2.14 0.00 0.00 1.43 0.00 0.00 0.82 0.00 0.00 1.28
(c) Dienes 3.78 5.41 0.03 0.00 0.55 0.41 6.04 11.86 0.00 0.00 0.00 0.00 4.23 4.35 0.00 0.00 0.00 0.00 1.32 0.40 0.00 0.00 0.00 0.00 0.42 0.00 0.00 0.00 0.00 0.00 0.16 0.00 0.00 0.00 0.00 0.00 0.09 0.00
0.00 0.00 0.00 0.98 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.58 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.02 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.08 0.00 0.00 1.34 0.00 0.00 0.58 0.00 0.00 0.36 0.00 0.00 0.08
0.00 0.09 0.00 0.00 0.69 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.27 0.00 0.00 2.15 0.00 0.00 0.92 0.00 0.00 0.41 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.41 0.00 0.00 3.32 0.00 0.00 0.61 0.00 0.00 0.81 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
degradation for polypropylene, polystyrene, and the binary mixture, rate constants were obtained from the
4722
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001
Table 2. Representative Values of Kinetic and Thermodynamic Parameters for Reaction Types Incorporated into Binary Kinetic Coupling Model of Polystyrene and Polypropylene Degradation
reaction family chain fission carbon-hydrogen bond fission hydrogen abstraction chain-end β-scission midchain β-scission radical recombination disproportionationb,c radical addition 1,5-hydrogen transfer
frequency factor, Ai (s-1 or L mol-1 s-1) 7.98 × 1015 3.5 × 1015
intrinsic barrier, Eo,i (kcal mol-1) 2.3b 0.0
representative heat of reaction for polystyrene, ∆Hj (kcal mol-1)
activation energy for polystyrene (kcal mol-1)
representative heat of reaction for polypropylene, ∆Hj (kcal mol-1)
activation energy for polypropylene (kcal mol-1)
1.0 1.0
69.7 78.5
72.0 78.5
86.0 94.8
88.3 94.8
-3.1
12.4
-3.1
12.4
16.4b
21.1
16.7b
21.3
transfer coefficient, Ri
1.5 × 108
13.3
2.62 × 1012 a
12.9
0.70 (endo) 0.30 (exo) 0.5
2.62 × 1012 a
12.9
0.5
19.5
22.7
19.8
22.8
0.0
-69.7
2.3
-86.0
2.3
0.0 0.5 0.70 (endo) 0.30 (exo)
-19.5 -3.1
2.3 3.2 12.4
-19.8 -3.1
2.3 3.2 12.4
2.17 × 109 b 1.14 × 108 9.7 × 106 a 2.57 × 109 d
2.3b 2.3 12.9 13.3
a Frequency factors from Deady et al. (1993).32 b Parameters obtained from Fried (1995).33 c Disproportionation estimated to be 5% of recombination rate at 350 °C (Schreck et al., 1989).34 d Estimated from data obtained by Yang and Shibasaki (1998).35
literature when available (see references in ref 20), and unknown values were estimated using the EvansPolanyi relationship25
kij ) Ai exp
(
)
-(Eo,i + Ri∆Hj) RT
(3)
where the rate constant (kij) depends on the frequency factor (Ai), the intrinsic barrier (Eo,i), and the transfer coefficient (Ri) for a given reaction type i and the heat of reaction, ∆Hj, for a particular reaction. Representative values of Ai, Eo,i, and Ri are listed in Table 2 and are derived from the polystyrene modeling study of Kruse et al.20 The first six columns of Table 2 are identical to Table 1 in their work. Judging from the mechanism in Figure 4, it is clear that the same reaction families govern both polypropylene and polystyrene decomposition. We assumed for this first-order analysis that the Ai, Eo,i, and Ri values are the same for polypropylene and polystyrene, consistent with the reaction family concept. However, the activation energies will be different for polypropylene and polystyrene for a given reaction family because the heats of reaction are different. The heats of reaction for polystyrene were obtained from Kruse et al.,20 and the heats of reaction for representative polypropylene and cross reactions were calculated from the NIST Structures and Properties Database26 based on relative radical stabilities. Using this rate constant information, the binary kinetic coupling model predicted that the rate enhancement for polystyrene was very close to unity, whereas that for polypropylene was much smaller than 1. This result suggests that the overall conversion for polystyrene should remain unchanged in the presence of polypropylene, whereas polypropylene degradation should be strongly retarded in the binary system. This result is consistent with the fact that the tertiary polystyrenederived radicals are more stable than tertiary polypropylene-derived radicals by 16.3 kcal/mol. If the two polymers are well-mixed, the polystyrene-derived radicals prefer self-interactions, and polypropylene-derived radicals preferentially abstract hydrogen from polystyrene instead of from polypropylene. Clearly, the predicted retardation of polypropylene degradation by the model is in contradiction with the experimental results.
The model developed by LaMarca et al. was based on a binary component system that is homogeneous, i.e., the two components are well-mixed. In polymer systems, the miscibility of two polymers depends on the Gibbs free energy of mixing. A negative free energy of mixing is one of the necessary conditions for miscibility.27,28 Several researchers have shown that polypropylene/ polystyrene blends are totally immiscible at low temperatures.29-31 Because the entropy of mixing is negligible for high-molecular-weight polymers with degrees of polymerization greater than 100,28 the polymers will also be immiscible at the temperature investigated here. Therefore, a homogeneous model is an inadequate description of this polypropylene/polystyrene binary mixture pyrolysis. This analysis suggests that other interactions are responsible for the observed enhancement in the degradation rate of polypropylene. As evidenced by the neat experimental data and supported by the lower carboncarbon bond dissociation energy for polystyrene (69.7 kcal/mol) than for polypropylene (86.0 kcal/mol), polystyrene degrades much more rapidly than polypropylene. It is possible that low-molecular-weight polystyrenederived radicals near the polymer interface diffuse into the polypropylene region and initiate the more recalcitrant polypropylene. Detailed kinetic modeling is currently in progress to explore this possibility further. Conclusions Total conversion of polypropylene during neat pyrolysis in a closed batch reactor reached around 60% at 420 °C and less than 4% at 350 °C. There was a significant influence of reaction temperature but only slight effects of sample loading. Four classes of products appeared during polypropylene pyrolysissalkanes, alkenes, dienes, and aromatic compoundsswith the first two present in significantly greater yields than the latter two. The product distribution also showed that most alkenes appeared in the form of C3n, alkanes in the form of C3n-1, and dienes in the form of C3n-2. This product distribution can be explained by the typical free-radical mechanism involving bond fission, hydrogen abstraction, β-scission, intramolecular hydrogen transfer, and radical recombination as the dominant steps.
Ind. Eng. Chem. Res., Vol. 40, No. 22, 2001 4723
For neat polystyrene pyrolysis, the polymer degraded completely in 10 min at 420 °C, and condensation products were observed at longer times. At 350 °C, the conversion to low-molecular-weight products reached approximately 75%. The major products of polystyrene pyrolysis were toluene, ethylbenzene, and styrene at 420 °C and styrene and 2,4-diphenyl-1-butene (styrene dimer) at 350 °C. During binary mixture pyrolysis of the two polymers, the overall conversion was higher than the average of the conversions of the neat cases at both temperatures. The conversion of polystyrene in binary reactions was similar to that in the neat cases. However, the conversion of polypropylene in binary reactions was enhanced by the presence of polystyrene. The enhancement is likely due to increased initiation through hydrogen abstraction from polypropylene by small polystyrenederived radicals, which diffuse into the polypropylene region as polystyrene degrades separately. Acknowledgment The authors are grateful for financial support from the Department of Energy (DE-FG22-96-PC96204), the CAREER Program of the National Science Foundation (CTS-9623741), and the MRSEC program of the National Science Foundation (DMR-9632472) at the Materials Research Center of Northwestern University. Literature Cited (1) Franklin Associates. Characterization of Municipal Solid Waste in the United States: 1998 Update; U.S. Environmental Protection Agency, U.S. Government Printing Office: Washington, D.C., 1999. (2) Rowatt, R. J. The Plastic Waste Problem. CHEMTECH 1993, 23, 56-60. (3) Shelley, S.; Fouhy, K.; Moore, S. Plastics Reborn. Chem. Eng. 1992, July, 30-35. (4) Davis, T. E.; Tobias, R. L.; Peterli, E. B. Thermal Degradation of Polypropylene. J. Polym. Sci. 1962, 56, 485-499. (5) Tsuchiya, Y.; Sumi, K. Thermal Decomposition Products of Polypropylene. J. Polym. Sci. A-1 1969, 7, 1599-1607. (6) Sousa Pessoa de Amorim, M. T.; Comel, C.; Vermande, P. Pyrolysis of Polypropylene: I. Identification of Compounds and Degradation Reactions. J. Anal. Appl. Pyrolysis 1982, 4, 73-81. (7) Chan, J. H.; Balke, S. T. The Thermal Degradation Kinetics of Polypropylene: Part I. Molecular Weight Distribution. Polym. Degrad. Stab. 1997, 57, 113-125. (8) Bockhorn, H.; Hornung, A.; Hornung, U.; Schawaller, D. Kinetic Study on the Thermal Degradation of Polypropylene and Polyethylene. J. Anal. Appl. Pyrolysis 1999, 48, 93-109. (9) Yamaguchi, T.; Watanabe, S.; Shimada, Y. Recovery of Monomeric Styrene from Polystyrene. Chemosphere 1973, 1, 7-10. (10) Ogawa, T.; Kuroki, T.; Ide, S.; Ikemura, T. Recovery of Indan Derivatives from Polystyrene Waste. J. Appl. Polym. Sci. 1981, 27, 857-869. (11) Ide, S.; Ogawa, T.; Kuroki, T.; Ikemura, T. Controlled Degradation of Polystyrene. J. Appl. Polym. Sci. 1984, 29, 25612571. (12) Ohtani, H.; Yuyama, T.; Tsuge, S.; Plage, B.; Schulten, H. R. Study on Thermal Degradation of Polystyrenes by PyrolysisGas Chromatography and Pyrolysis-Field Ionization Mass Spectrometry. Eur. Polym. J. 1990, 26 (8), 893-899. (13) Gardner, P.; Lehrle, R. Polystyrene Pyrolysis Mechanisms I. As Deduced from the Dependence of Product Yields on Film Thickness. Eur. Polym. J. 1993, 29, 425-435.
(14) McCaffrey, W. C.; Brues, M. J.; Cooper, D. G.; Kamal, M. R. Thermolysis of Polyethylene/Polystyrene Mixtures. J. Appl. Polym. Sci. 1996, 60, 2133-2140. (15) Ibrahim, M. M.; Hopkins, E.; Seehra, M. S. Thermal and Catalytic Degradation of Commingled Plastics. Fuel Process. Technol. 1996, 49, 65-73. (16) Woo, O. S.; Broadbelt, L. J. Recovery of High-Value Products from Styrene-Based Polymers through Coprocessing: Experiments and Mechanistic Modeling. Catal. Today 1998, 40, 121-140. (17) Madras, G.; McCoy, B. J. Distribution Kinetics for Polymer Mixture Degradation. Ind. Eng. Chem. Res. 1999, 38, 352-357. (18) Williams, P. T.; Williams, E. A. Interaction of Plastics in Mixed-Plastics Pyrolysis. Energy Fuels 1999, 13, 188-196. (19) Woo, O. S.; Ayala, N.; Broadbelt, L. J. Mechanistic Interpretation of Base-Catalyzed Depolymerization of Polystyrene. Catal. Today 2000, 55, 161-171. (20) Kruse, T. M.; Woo, O. S.; Broadbelt, L. J. Detailed Mechanistic Modeling of Polymer Degradation: Application to Polystyrene. Chem. Eng. Sci. 2001, 56, 971-979. (21) De Witt, M. J.; Broadbelt, L. J. Binary Interactions between High-Density Polyethylene and 4-(1-Naphthylmethyl)bibenzyl during Low-Pressure Pyrolysis. Energy Fuels 2000, 14 (2), 448-458. (22) Dotson, N. A.; Galvan, R.; Laurence, R. L.; Tirrell, M. Polymerization Process Modeling; John Wiley and Sons: New York, 1996; pp 344-345. (23) Uemichi, Y.; Kashiwaya, Y.; Tsukidate, M.; Ayame, A.; Kanoh, H. Product Distribution in Degradation of Polypropylene Over Silica-Alumina and CaX Zeolite Catalysts. Bull. Chem. Soc. Jpn. 1983, 56, 2768-2773. (24) LaMarca, C.; Libanati, C.; Klein, M. T. Design of Kinetically Coupled Complex Reaction Systems. Chem. Eng. Sci. 1990, 45, 2059-2065. (25) Evans, M. G.; Polanyi, M. Inertia and Driving Force of Chemical Reactions. Trans. Faraday Soc. 1938, 34, 11-29. (26) Stein, S. E.; Lias, S. G.; Liebman, J. F.; Kafafi, S. A. NIST Structure and Properties Users Guide; National Institute of Standards and Technology: Gaithersburg, MD, 1994. (27) Paul, D. R.; Barlow, J. W. Polymer Blends (or Alloys). J. Macrmol. Sci.-Rev. Macromol. Chem. 1980, C18 (1), 109-168. (28) Eastmond, G. C. In Advances in Polymer Science: Biomedical Applications/Polymer Blends; Springer: New York, 1999; Vol. 149, pp 69-79. (29) Krause, S. In Polymer Blends; Paul, D. R.; Newman, S., Eds.; Academic Press: New York, 1978; Vol. 1, p 74. (30) Radonjic, G. Compatibilization Effects of Styrenic/Rubber Block Copolymer in Polypropylene/Polystyrene Blends. J. Appl. Polym. Sci. 1999, 72, 291-307. (31) Fortelny, I.; Michalkova, D.; Mikesova, J. Effect of the Mixing Conditions on the Phase Structure of PP/PS Blends. J. Appl. Polym. Sci. 1996, 59, 155-164. (32) Deady, M.; Mau, A. W. H.; Moad, G.; Spurling, T. H. Evaluation of the Kinetic Parameters for Styrene Polymerization and Their Chain Length Dependence by Kinetic Simulation and Pulsed Laser Photolysis. Makromol. Chem. 1993, 194, 1691-1705. (33) Fried, J. R. Polymer Science and Technology; Prentice Hall: Upper Saddle River, NJ, 1995; pp 33-38. (34) Schreck, V. A.; Serelis, A. K.; Solomon, D. H. Self-Reactions of 1,3-Diphenylpropyl and 1,3,5-Triphenylpentyl RadicalssModels for Termination in Styrene Polymerization. Aust. J. Chemistry 1989, 42, 375-393. (35) Yang, M.; Shibasaki, Y. Mechanisms of Thermal Degradation of Polystyrene, Polymethacrylonitrile, and Their Copolymers on Flash Pyrolysis. J. Polym. Sci. A: Polym. Chem. 1998, 36, 2315-2330.
Received for review February 21, 2001 Revised manuscript received July 20, 2001 Accepted August 12, 2001 IE010171S
